Programmable address processor for graphics applications
a graphics application and address processor technology, applied in the field of computer systems, can solve the problems of restricting performance, severely restricting the performance of the gpu, and the cost associated with memory lookups remains relatively high when compared to the cost of processing tim
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Benefits of technology
Problems solved by technology
Method used
Image
Examples
example method embodiments
[0090]FIG. 7 shows flowchart 700 providing example steps for processing memory lookup requests, according to an embodiment of the present invention. Other structural and operational embodiments will be apparent to persons skilled in the relevant art(s) based on the following discussion. Flowchart 700 is described with reference to the embodiment of FIG. 3. However, flowchart 700 is not limited to that embodiment. The steps shown in FIG. 7 do not necessarily have to occur in the order shown. The steps of FIG. 7 are described in detail below.
[0091]Flowchart 700 begins with step 702. In step 702, a primary memory lookup request is received. For example, in FIG. 3, input module 302 receives a memory lookup request from L1 cache 106.
[0092]In step 704, the primary memory lookup request is classified. For example, in FIG. 3, input module 302 can classify a received primary memory lookup request as a first memory lookup type. For example, the first memory lookup type can be a linked list tr...
example graphics processing embodiments
[0118]Aspects of the present invention may be applied in graphical processing units to increase the efficiency of memory lookups. For example, the methods and systems described herein may be applied to data structure traversals that may involved when dealing with vector textures (e.g., font rendering), texture trees (e.g., wavelet decompression), isosurface extraction, grid computations involved in graph processing (e.g., fast fluid dynamics), sparse matrices as applied in physics and elsewhere, and other tasks as would be appreciated by those skilled in relevant art(s).
[0119]In the above mentioned examples, linked list, trees, and combinations thereof are traversed in a series of dependent memory lookups with a series of computations that involve relatively simple computations (e.g., comparisons), and thus may be ideal for an APU with limited computational resources. The results of the traversals are often reused by elements of a GPU.
[0120]For example, FIG. 9A is an illustration of...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


